Acute myeloid leukemia (AML) is the most common type of adult leukemia and second most common form of childhood leukemia. Some patients have a good chance of cure because historically we know that if they have certain chromosomes that are altered, those patients have done well in response to chemotherapy. The opposite case is also true. More recently we have been able to find mutations in certain genes that give these same good or bad chances for cure to patients with AML. The human gene called FLT3 was cloned by my lab about 20 years ago. It turns out to be the most frequently mutated gene in AML. Not only is it so frequently mutated, but the most common type of FLT3 mutation (called ITD) also gives a very aggressive leukemia with a horrible chance of cure for the AML patients who have the mutation. For example, in pediatric AML, patients without the mutation have a 50-60% chance of cure but those with a FLT3/ITD mutation have only a 15-20% chance of cure. Thus, to improve the chance for curing these patients we need to find ways to reverse the very lethal aggressive nature imparted to the leukemia by way that FLT3 signals in the cell. One of the ways we have attacked this problem was to find the first drugs that were able to block how c signals. This is called a tyrosine kinase inhibitor or TKI. We showed that this would preferentially kill leukemia cells with the FLT3/ITD mutation while leaving normal cells alone. Later generations of these FLT3 TKI are in advanced clinical trials to try to improve the cure rate for FLT3/ITD AML patients. Another type of FLT3 mutation, called a kinase domain or KD mutation, does not give patients with AML a worse chance for cure. This gives us the opportunity to try to learn how the 2 different kinds of mutations in the same gene can lead to such different outcomes. If we can understand how the ITD vs. KD mutations of FLT3 signal differently, it should point out the pathway that results in really bad, difficult to cure leukemias. These same pathways are likely to be used by other leukemias and possibly other types of cancers and so identifying them will be the first step followed by targeting them to improve the cure rate for these diseases. Because patient leukemias have combinations of so many different types of mutations, it is not possible to sort out the signaling differences between them that are due to FLT3/ITD vs. FLT3/KD mutations. To overcome this problem we have generated mice in which we genetically engineered them to be born with either of the two types of mutations. These mice are genetically identical other than the type of FLT3 mutations we have engineered so when we combine them with the same second hits required to generate leukemia any differences between the leukemias are a result of differences in how the FLT3 mutations function. This will enable us to determine what gives the ITD mutations its bad characteristics. We can develop targeted therapies for this bad pathway that is likely used by other difficult to cure leukemias and perhaps other cancers.